← Back to Skills Marketplace

Skill

Name: Skill
Author: paulpreibisch

by Paul Preibisch · GitHub ↗ · v4.6.6 · MIT-0

cross-platform ✓ Security Clean

1752

Downloads

Stars

Active Installs

Versions

Install in OpenClaw

/install agentvibes-openclaw-skill

Description

🎤 AgentVibes TTS for Claude Code & OpenClaw — Switch voices, set personality, control speed, background music, language learning mode, reverb/effects, and m...

README (SKILL.md)

🎤 AgentVibes Voice Management

Professional text-to-speech for Claude Code and OpenClaw. Free, offline, no account required.

Providers: Piper TTS (914+ voices, all platforms) · macOS Say (built-in) · Windows SAPI (zero setup) · Soprano (neural)

Voice Commands

/agent-vibes:switch \x3Cvoice_name>

Switch to a different voice.

/agent-vibes:switch en_US-amy-medium
/agent-vibes:switch en_GB-alan-medium
/agent-vibes:switch fr_FR-siwis-medium

/agent-vibes:list [first|last] [N]

List available voices.

/agent-vibes:list                    # Show all voices
/agent-vibes:list first 5            # Show first 5
/agent-vibes:list last 3             # Show last 3

/agent-vibes:preview [first|last] [N]

Preview voices with audio samples.

/agent-vibes:preview                 # Preview first 3 voices
/agent-vibes:preview 5               # Preview first 5
/agent-vibes:preview last 5          # Preview last 5

/agent-vibes:sample \x3Cvoice_name>

Play a sample of a specific voice.

/agent-vibes:sample en_US-ryan-high

/agent-vibes:get

Show the currently active voice.

/agent-vibes:set-favorite-voice

Mark current voice as your favorite.

Personality & Style

/agent-vibes:personality [name|list|add|edit|get|reset]

Set a personality style for TTS output.

/agent-vibes:personality list          # Show available personalities
/agent-vibes:personality sarcastic     # Switch to sarcastic style
/agent-vibes:personality dramatic      # Switch to dramatic style
/agent-vibes:personality reset         # Back to default

/agent-vibes:set-pretext \x3Cphrase>

Add a spoken prefix before every TTS message.

/agent-vibes:set-pretext "AgentVibes"   # Speaks "AgentVibes: ..." before each message
/agent-vibes:set-pretext ""              # Clear pretext

Speed & Effects

/agent-vibes:set-speed \x3Cspeed>

Control speech rate (0.5x – 3.0x).

/agent-vibes:set-speed 1.0             # Normal speed
/agent-vibes:set-speed 1.5             # 50% faster
/agent-vibes:set-speed 0.8             # Slower

/agent-vibes:effects [reverb|echo|pitch|eq|reset]

Configure voice effects.

/agent-vibes:effects reverb hall       # Hall reverb
/agent-vibes:effects reverb none       # No reverb
/agent-vibes:effects reset             # Clear all effects

Background Music

/agent-vibes:background-music [on|off|status|list|switch]

Toggle or change background music played under TTS.

/agent-vibes:background-music on       # Enable background music
/agent-vibes:background-music off      # Disable
/agent-vibes:background-music list     # Show available tracks
/agent-vibes:background-music switch jazz  # Switch to jazz track

Verbosity

/agent-vibes:verbosity [low|medium|high]

Control how much Claude speaks while working.

/agent-vibes:verbosity low             # Brief acknowledgments only
/agent-vibes:verbosity medium          # Key decisions (default)
/agent-vibes:verbosity high            # Full reasoning

Mute / Replay

/agent-vibes:mute / /agent-vibes:unmute

Silence or restore TTS output (persists across sessions).

/agent-vibes:replay [N]

Replay recent audio (last 10 kept).

/agent-vibes:replay                    # Replay last audio
/agent-vibes:replay 2                  # Replay second-to-last

Language & Learning

/agent-vibes:language \x3Clang>

Set your native language.

/agent-vibes:language english
/agent-vibes:language japanese

/agent-vibes:learn [on|off]

Enable language learning mode — Claude speaks in both your native and target language.

/agent-vibes:learn on
/agent-vibes:learn off

/agent-vibes:translate \x3Ctext>

Translate and speak text in the target language.

Provider Management

/agent-vibes:provider [list|switch|info]

/agent-vibes:provider list
/agent-vibes:provider switch piper     # Piper TTS (free, offline, 914+ voices)
/agent-vibes:provider switch macos     # macOS Say (Mac only)
/agent-vibes:provider switch sapi      # Windows SAPI (Windows only, zero setup)
/agent-vibes:provider switch soprano   # Soprano (neural)

Providers

Provider	Platform	Cost	Voices
Piper TTS	All platforms	Free, offline	914+ in 30+ languages
macOS Say	macOS only	Free (built-in)	100+ system voices
Windows SAPI	Windows only	Free (built-in)	System voices, zero setup
Soprano	All platforms	Free	Neural voices

Miscellaneous

/agent-vibes:whoami

Show current AgentVibes configuration.

/agent-vibes:version

Show installed version.

/agent-vibes:update

Update AgentVibes to the latest version.

/agent-vibes:show / /agent-vibes:hide

Show or hide the AgentVibes status indicator.

/agent-vibes:cleanup / /agent-vibes:clean

Remove cached audio files.

Default Voices (Piper TTS — Free & Offline)

English (US): en_US-lessac-medium · en_US-amy-medium · en_US-ryan-high · en_US-libritts-high (914 speakers)

English (UK): en_GB-alan-medium · en_GB-jenny_dioco-medium

French: fr_FR-siwis-medium · fr_FR-gilles-low

German: de_DE-thorsten-medium · de_DE-eva_k-x_low

Spanish: es_ES-davefx-medium · es_MX-claude-high

Japanese: ja_JP-ayanami-medium · Chinese: zh_CN-huayan-x_low · Korean: ko_KR-kss-medium

+ 900 more across 30+ languages. All voices are downloaded from HuggingFace — no account required.

Tips

Preview first: Use /agent-vibes:preview before committing to a voice
Verbosity: Set to low for focused work, high for full narration
BMAD party mode: Each agent gets their own voice, music, and personality
Replay: Use /agent-vibes:replay to re-hear the last 10 responses
Speed: Combine with personality for a fully custom TTS character

Enjoy your TTS experience! 🎵

Usage Guidance

This skill appears coherent with its TTS purpose and requests no credentials. Before installing, consider: 1) it likely needs network access and disk write permissions to download and cache Piper voice files (HuggingFace), which can be large; 2) the SKILL.md doesn't state where cached audio or voices are stored — if you have storage or privacy concerns, ask the developer for storage paths and retention policy; 3) verify any external provider (e.g., 'Soprano') if you plan to use it, since that may use online services; otherwise the skill's scope and requirements look proportionate.

Capability Assessment

✓ Purpose & Capability

The name/description (an offline TTS voice manager) matches the SKILL.md commands (switch voices, set personality, speed, effects, background music, replay, etc.). The mention of Piper voices hosted on HuggingFace fits the claimed capability.

✓ Instruction Scope

SKILL.md only defines TTS-related commands and user-facing behaviors. It does not instruct the agent to read unrelated files, environment variables, or secrets, nor to exfiltrate data. One minor ambiguity: it references downloading voices (HuggingFace) and persistent caches/replay, but does not specify storage locations or exact download steps.

✓ Install Mechanism

There is no install spec or code — instruction-only (lowest install surface). The only installation-like action implied is fetching voice assets from HuggingFace; no arbitrary third-party download URLs, installers, or extracted archives are specified in the SKILL.md.

✓ Credentials

The skill declares no required environment variables, credentials, or config paths. That is proportional to a local/offline TTS controller. (Note: runtime will still need network access to download voices if using Piper from HuggingFace, and will need write access to store caches.)

ℹ Persistence & Privilege

always is false and autonomous invocation is allowed (normal). The skill states it persists mute/voice favorites and caches recent audio; this implies local storage use but does not request elevated or cross-skill privileges. The persistence model is reasonable but unspecified (where/how long data is stored).

How to Use

Make sure OpenClaw is installed (local or Docker)
Run the install command in chat: /install agentvibes-openclaw-skill
After installation, invoke the skill by name or use /agentvibes-openclaw-skill
Provide required inputs per the skill's parameter spec and get structured output

Version History

v4.6.6

914+ voices, personality styles, speed control, background music, reverb/effects, language learning mode, verbosity control, Windows SAPI + Soprano providers

v4.6.5

Big update: Full overhaul of docs to cover new cross-platform features, 900+ voices, and advanced voice controls. - Rewrote documentation for clarity and completeness, including new voice and provider commands. - Added support for Windows SAPI, Soprano neural voices, and over 900+ Piper voices. - Introduced commands for personality, background music, speech speed, effects, verbosity, language learning, and more. - Expanded examples and provider comparison table for easy setup on all platforms. - Updated tips and quick references to reflect enhanced capabilities and new features.

v3.3.0

Initial ClawHub release. Stream free TTS from voiceless servers to Linux/macOS/Android. Two deployment options: server-side TTS + audio streaming OR efficient text streaming + receiver-side generation.

Metadata

Slug agentvibes-openclaw-skill

Version 4.6.6

License MIT-0

All-time Installs 5

Active Installs 5

Total Versions 3

Frequently Asked Questions

What is Skill?

🎤 AgentVibes TTS for Claude Code & OpenClaw — Switch voices, set personality, control speed, background music, language learning mode, reverb/effects, and m... It is an AI Agent Skill for Claude Code / OpenClaw, with 1752 downloads so far.

How do I install Skill?

Run "/install agentvibes-openclaw-skill" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.

Is Skill free?

Yes, Skill is completely free, licensed under MIT-0. You can download, install and use it at no cost.

Which platforms does Skill support?

Skill is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).

Who created Skill?

It is built and maintained by Paul Preibisch (@paulpreibisch); the current version is v4.6.6.

More Skills